Aminoguanidine hydrazone derivatives as non-peptide NPFF1 receptor antagonists reverse opioid induced hyperalgesia

Article abstract of DOI:10.1021/acschemneuro.8b00099

Neuropeptide FF receptors (NPFF1R and NPFF2R) and their endogenous ligand Neuropeptide FF have been shown previously to display anti-opioid properties and to play a critical role in the adverse effects associated with chronic administrations of opiates including the development of opioid-induced hyperalgesia and analgesic tolerance. In this work, we sought to identify novel NPFF receptors ligands by focusing our interest on a series of heterocycles as rigidified non-peptide NPFF receptor ligands, starting from already described aminoguanidine hydrazones (AGH's). Binding experiments and functional assays highlighted AGH 1n and its rigidified analog 2-amino-dihydropyrimidine 22e for in vivo experiments. As earlier shown with the prototypical dipeptide antagonist RF9, both 1n and 22e reduced significantly the long lasting fentanyl-induced hyperalgesia in rodents. Altogether these data indicate that AGH rigidification maintains nanomolar affinities for both NPFF receptors, while improving antagonist character towards NPFF1R.

Full text of DOI:10.1021/acschemneuro.8b00099

Subscriber access provided by Kaohsiung Medical University
Article
Aminoguanidine hydrazone derivatives as non-peptide NPFF1
receptor antagonists reverse opioid induced hyperalgesia
Hassan Hammoud, Khadija Elhabazi, Raphaëlle Quillet, Isabelle Bertin, Valérie Utard, Emilie Laboureyras,
Jean-Jacques Bourguignon, Frederic Bihel, Guy Simonnet, Frederic Simonin, and Martine Schmitt
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00099 • Publication Date (Web): 04 May 2018
Downloaded from http://pubs.acs.org on May 8, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Aminoguanidine hydrazone derivatives as nonꢀ
peptide NPFF1 receptor antagonists reverse opioid
induced hyperalgesia
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hassan Hammoud,a Khadija Elhabazi,b Raphäelle Quillet, b Isabelle Bertin,b Valérie
Utard,b Emilie Laboureyras,c, Jean-Jacques Bourguignon,a Frédéric Bihel,a Guy Simonnet,c
Frédéric Simonin,b*# Martine Schmitt.a*#
aUniversity of Strasbourg, CNRS, UMR7200, Faculty of Pharmacy, Fꢀ67401 Illkirch
Graffenstaden
bUniversité de Strasbourg, CNRS, Biotechnologie et Signalisation Cellulaire, UMR 7242, Fꢀ
67401 Illkirch Graffenstaden
c HoméostasieꢀAllostasieꢀPathologieꢀRéhabilitation, UMR 5287 CNRS, Université de
Bordeaux, Bordeaux, France.
KEYWORDS: GPCR, neuropeptide FF, GPR74, aminoguanidine hydrazone, Opioidꢀinduced
hyperalgesia.
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 33
1
2
3
ABSTRACT
4
5
6
7
8
9
Neuropeptide FF receptors (NPFF1R and NPFF2R) and their endogenous ligand
Neuropeptide FF have been shown previously to display antiꢀopioid properties and to play a
critical role in the adverse effects associated with chronic administrations of opiates including
the development of opioidꢀinduced hyperalgesia and analgesic tolerance. In this work, we
sought to identify novel NPFF receptors ligands by focusing our interest on a series of
heterocycles as rigidified nonꢀpeptide NPFF receptor ligands, starting from already described
aminoguanidine hydrazones (AGH’s). Binding experiments and functional assays highlighted
AGH 1n and its rigidified analog 2ꢀaminoꢀdihydropyrimidine 22e for in vivo experiments. As
earlier shown with the prototypical dipeptide antagonist RF9, both 1n and 22e reduced
significantly the long lasting fentanylꢀinduced hyperalgesia in rodents. Altogether these data
indicate that AGH rigidification maintains nanomolar affinities for both NPFF receptors,
while improving antagonist character towards NPFF1R.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 33
ACS Chemical Neuroscience
1
2
3
INTRODUCTION
4
5
6
7
8
Neuropeptide FF (NPFF) is an endogenous octapeptide member of a family of neuropeptides
that display a conserved ArgꢀPheꢀNH2 at their Cꢀterminal end and have therefore been named
RFꢀamide peptides. It binds to specific Gꢀproteinꢀcoupledꢀreceptors (GPCRs) that exist as two
different subꢀtypes, NPFF1R (GPR147) and NPFF2R (GPR74).1ꢀ4 These two receptors have
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
been characterized in mammals including man. They display different cellular and tissue
3, 5
distributions in the central nervous system and at the periphery.
NPFF2R are mostly coꢀ
localized with µꢀopioid receptors, in particular in the spinal cord, whereas NPFF1R are
mainly present in the brain. The physiological roles of these systems are still poorly
understood. However they have been involved in several physiological processes, such as
insulin release, food intake, 6ꢀ8 memory, 9ꢀ 11 blood pressure 12 and electrolyte balance.13 More
interestingly, several lines of evidence support the involvement of NPFF receptors in the
modulation of nociception and opioidꢀinduced analgesia. 14
Long term opioid treatments are associated with several adverse effects, including the
development of analgesic tolerance,15 which leads to the need of increasing doses of the
opioid analgesic to be used for relieving pain.16ꢀ17 It has been proposed that this phenomenon
may involve a balance between opioid and soꢀcalled antiꢀopioid systems able to modulate
nociception. More specifically, stimulation of muꢀopioid receptors triggers activation of antiꢀ
opioid systems such as NPFF,14, 18 that in turn produce hyperalgesia 19ꢀ21 thus diminishing the
net analgesic effect of the opioid analgesic.
A better understanding of the respective physiological roles played by both NPFF1R and
NPFF2R in the modulation of nociception is limited by the lack of pharmacological tools,
including agonists (metabolically stable NPFF mimetics) and antagonists acting in a selective
manner on one of the two NPFF receptor subtypes. In our previous work, we focused our
efforts on the prototypical signature of the Cꢀterminal part of NPFF (ArgꢀPheꢀNH2) and
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 33
1
2
3
4
5
6
7
8
9
reported the first nonꢀselective NPFF receptor antagonist (RF9, Nꢀadamantaneꢀcarbonyl Lꢀ
Arg LꢀPhe CONH2). 22, 24 We further showed that RF9 did not show any intrinsic analgesic or
hyperalgesic effect but efficiently blocks hyperalgesia induced by repeated administration of
opiates such as heroin and morphine as well as analgesic tolerance. 22, 24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The very few SAR analyses published so far in the literature support the critical role played
by the guanidinium cation present in the structure of Nꢀacyl arginineꢀcontaining dipeptides
recently described by us, 25ꢀ 26 or in the few reported nonpeptide structures (Fig. 1, guanidineꢀ
piperidines I, heteroaryl guanidines II and aminoguanidine hydrazones III). 27, 29
Figure 1. Guanidineꢀcontaining nonpeptide NPFFꢀR ligands
AGHs were initially reported by Acadia pharmaceuticals, but only very few derivatives were
23, 29ꢀ30
described in terms of binding and functional activities.
As AGHs are versatile
pharmacological tools, compatible with in vivo oral administration (ex: Guanabenz, an
antihypertensive drug), we describe in this paper an inꢀdepth SAR analysis of the AGH
moiety, as well as a scaffold hopping strategy (design and synthesis of semiꢀconstrained AGH
analogues) leading to promising heterocyclic structures with in vivo efficacy against opioidꢀ
induced hyperalgesia. Formally they replace the prototypical set of H bond interactions by a
ring system able to mimic the AGH chemotype in a specific conformation (see Fig. 2).
ACS Paragon Plus Environment

Page 5 of 33
ACS Chemical Neuroscience
1
2
3
Figure 2. Scaffold hopping strategy along the structure of AGH chemotype.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CHEMISTRY
The different aminoguanidine arylhydrazones 1 could be easily obtained by reacting the
corresponding aryl or heteroaryl carboxaldehyde with amino guanidine hydrochloride in
refluxing ethanol or microwaves, when the carboxaldehyde was less reactive (yields ranging
from 75 to 90%).31ꢀ 32 Most of aldehydes were commercially available (cpds 1aꢀe, 1g-p), or
could be prepared by a SuzukiꢀMiyaura reaction under microwaves (cpds 1f, 1q-s), as
depicted in Scheme 1.
All the synthesized aryl and heteroaryl aminoguanidine hydrazones are listed in Tables 1ꢀ2.
Scheme 1. Synthesis of AGH’s 1
Reagents and conditions: (a) ArB(OH)2, Pd(PPh3)4,
Na2CO3, DME/H2O, MW, 150°C, 89ꢀ97%, 10min; (b))
Amino guanidine, HCl, EtOH, reflux, 5h, 75ꢀ90%.
The guanidine fragment constitutes a second anchor point allowing further introduction of
various substituents at either the terminal (ω), or hydrazine (α) nitrogen. The substitution on ω
position was performed as indicated in Scheme 2. Chemical diversity (NHR) could be
introduced either at the final step (method A), or at the beginning (method B). Both are
yielding similar cumulative yields. However, with fairly active nucleophiles (anilines), the
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 33
1
2
3
4
5
method B involving sodium thiocyanate was found more efficient because of its strong
electrophilic character.
6
7
8
Scheme 2. Substitution on ω position of the aminoguanidines hydrazones
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (a) ArCHO, EtOH, reflux, 5h; (b) Iodomethane, DMF, rt, 4h; (c) NH2R, EtOH,
reflux, 5h; or NH2R, EtOH, MW, 120°C, 15min; (d) Sodium thiocyanate, EtOH, reflux, 30 min. (e)
Iodomethane, DMF, rt; (f) Hydrazine monohydrate, MeOH, reflux, 3h.
The thiosemicarbazone 2, obtained by refluxing the title carboxaldehyde with
thiosemicarbazide, was quantitatively methylated by means of iodomethane in DMF. The
resulting Sꢀmethyl intermediate 3 could be reacted with various primary and secondary
amines (method A).33ꢀ34 In the method B, the Nꢀphenyl thiourea intermediate 5 was further
methylated. The resulting Sꢀmethyl thiourea derivative 6 was reacted with hydrazine and the
final Nꢀaryl aminoguanidine hydrazone 4f could be prepared in satisfactory yield.35ꢀ36
Direct substitution of the highly nucleophilic aminoguanidine hydrazone was performed in
various solvents, bases and halides (method A, Scheme 3). The reaction of 1n with benzyl
bromide in DMF was efficient and regioselective, leading to the Nꢀαꢀbenzyl derivative 8 as
the single isomer. However alkylation with highly reactive methyl iodide afforded a mixture
of mono and diꢀmethyl derivatives, whereas no reaction occurred with the less electrophilic
phenethyl bromide, even in DMF solution. An alternative method (method B) could be
generalized to all derivatives. Sꢀmethyl thiourea was reacted with Nꢀalkylhydrazine in ethanol
leading in a nonꢀambiguous manner to the Nꢀalkyl aminoguanidine intermediate 9. All the
synthesized Nꢀsubstituted aminoguanidine hydrazones are summarized in Table 3.
ACS Paragon Plus Environment

Page 7 of 33
ACS Chemical Neuroscience
1
2
3
Scheme 3. Substitution at the nonꢀterminal nitrogen (AGH’S 8)
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (a) BnBr, DMF or
MeOH, rt, 48h; (b) NH2ꢀNHR1, EtOH, reflux; (c)
3,4ꢀdichlorobenzaldehyde, EtOH, reflux.
The guanidine moiety was also introduced in various cyclic systems, as depicted in Scheme 4.
Cyclic thioureas were first Sꢀmethylated, and the resulting intermediates 10 were reacted with
hydrazine affording the 2ꢀhydrazino cyclic amidines 11 as the key intermediate (Scheme 4).
Scheme 4. Access to the cyclic aminoguanidine derivatives 12 and 13.
Reagents and conditions: (a) MeI, MeOH, reflux,
1h; (b) Hydrazine monohydrate, EtOH, reflux, 6h;
(c) ArCHO, EtOH, reflux, 5h.
Other AGH cyclic derivatives bridging the benzaldehyde imine carbon to the terminal
guanidine nitrogen were also synthesized. The 5ꢀarylꢀ2ꢀaminothiadiazines 14 a-c were
prepared in one step by reacting thiosemicarbazide with a substituted phenacyl bromide in
methanol.37 In a similar way the different dihydrotriazines 15 could be obtained by replacing
thiosemicarbazide by aminoguanidine in cyclocondensation reaction. As shown in Scheme 5,
38
the 6ꢀmethyl dihydrotriazine 15 could be recovered in good yield,
whereas the reaction
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 33
1
2
3
4
5
6
7
8
9
performed with phenacyl chloride yielded a mixture of dihydrotriazine 16 as the minor
product along with the diꢀadduct 17. As the expected compound 16 could not be isolated from
the reaction mixture, a second approach was carried out with success.
Scheme 5. Rapid access to 5ꢀarylꢀ2ꢀaminothiadiazines 14 and 6ꢀmethylꢀ3ꢀ
aminodihydrotriazine 15 .
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
S
S
N
Br
Ar
H2N
H2N
Ar
R
NHR
NH2
N
H
NHR
a
N
N
14
O
NH
NH
NH2
Cl
R
N
H
N
b or c
R= Me, 15, major
R = Ph, 16, minor
+
O
NH
H
N
R
R = Me , not observed
R = Ph, 17, major
R
N
N
H
H
O
Reagents and conditions: (a) MeOH, reflux,
30min, 48ꢀ86%; (b) H2O, 25°C, 48h, 93%; (c) H2O,
25°C; or EtOH, reflux
.
The key step was a careful reduction of the corresponding 2ꢀaminoꢀ1,2,4ꢀtriazines 20a and
39
20b into the corresponding dihydrotriazines 16a-b by means of triethylsilane in TFA,
avoiding removal of critical chlorine atoms from the phenyl ring observed when using
catalytic hydrogenation. The 2ꢀaminoꢀ1,2,4ꢀtriazines 18 were prepared by cyclocondensation
of different aryl glyoxals with ethyl hydrazinocarbimidothioate leading to a mixture of two
regioisomers 18 (major) and 19 (minor, <10%), respectively.40 They were easily recovered by
silica gel column chromatography (Scheme 6).
ACS Paragon Plus Environment

Page 9 of 33
ACS Chemical Neuroscience
1
2
3
Scheme 6. Access to 6ꢀarylꢀ3ꢀ aminoꢀdihydrotriazines 16a, b
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Reagents and conditions: (a) EtOH, reflux, 30
min; (b) 7N NH3 in MeOH, reflux, 15h; (c) H2/Pd,
60psi, 15h; (d) Et3SiH, TFA, 15h, rt.
In a similar manner reduction of 5ꢀarylꢀ2ꢀaminopyrimidines 21a-e, obtained by reacting the
41
commercially available 2ꢀaminoꢀ5ꢀiodopyrimidine with different arylboronic acids,
afforded the corresponding dihydropyrimidines 22a-e (Scheme 7).
Scheme 7. Access to 5ꢀarylꢀ2ꢀaminoꢀdihydropyrimidines 22a-e and 25a-c.
Reagents and conditions: (a) ArB(OH)2,
Pd(OAc)2, SꢀPhos, K2CO3, MeCN, H2O, 95°C, 15h;
(b) Et3SiH, TFA, rt, 1h; (c) ArB(OH)2, Pd(OAc)2,
KF, MeOH, µw, 120°C; (d) NH2R2, THF, µw,
100°C, 12min
The use of umpolung principle allowed rapid access to 2ꢀNꢀsubstituted 2ꢀamino pyrimidine
derivatives 24, as illustrated in Scheme 7. The 5ꢀarylꢀ2ꢀchloropyrimidine 23 was submitted to
amination using microwaves. The resulting derivatives 24a-c were reduced as described
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 33
1
2
3
4
5
39
earlier, giving the expected dihydropyrimidines 25a-c in satisfactory yields.
Compounds 14, 16, 22 and 25 are listed in Table 5.
Cyclic
6
7
8
RESULTS AND DISCUSSION
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Binding experiments were performed first by using [3H]ꢀFFRFꢀNH2 as radioligand on
membranes from CHO cells stably expressing SFꢀhNPFFR1 or SFꢀhNPFFR2.42 The
percentages of inhibition of FFRFꢀNH2 binding by the different compounds at 0.5 and 5 µM
are shown in Tables 1ꢀ4. Ki values were determined for the most potent compounds.
Altogether our data allowed the following comments: i) The unsubstituted AGH 1a presented
a good affinity for NPFF1R (Ki= 76 nM), associated with a satisfactory selectivity towards
NPFF2R (Ki > 3 µM). ii) Surprisingly substitution in 4ꢀposition by chlorine (1d), or methoxy
(1j) afforded compounds with similar affinities towards NPFF1R (Ki ± 20 nM). iii)
Introduction of a second chlorine atom in position 3 (1n), or 2 (1l) gave potent ligands but
with different selectivity profiles. Compound 1n displayed a good affinity of NPFF1R (Ki =
12 nM) and was relatively selective toward NPFF2R (twentyfold), whereas 1l is not selective.
iv) Replacing the 3ꢀCl substituent in 1n (AC-1045) by a CF3 group dramatically reversed
selectivity profile. Finally, within this chemical subꢀseries the AGH 1p (AC-3099) displayed
the best affinity for NPFF2R (Ki = 4 nM) and a ten fold lower affinity for NPFF1R.
ACS Paragon Plus Environment

Page 11 of 33
ACS Chemical Neuroscience
1
2
3
Table 1. AGH derivatives and their binding affinities for NPFF1R & NPFF2R.
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NPFF1-R
NPFF2-R
Ki a
(nM)
Ki a
(nM)
% inhibition at
% inhibition at
nº
1a
1b
1c
1d
1e
1f
Ar
X
5µM
62
55
56
98
70
9
0.5µM
34
5µM
28
13
24
60
35
30
23
8
0.5µM
H
76
ndb
nd
20
nac
nd
nd
nd
nd
25
341
47
86
10
1
3,600
nd
2ꢀCl
25
3ꢀCl
17
0
nd
4ꢀCl
69
18
17
0
490
na
2ꢀPh
22
3ꢀPh
0
nd
1g
1h
1i
4ꢀPh
58
48
59
77
100
66
77
11
5
nd
2ꢀOMe
3ꢀOMe
4ꢀOMe
4ꢀCF3
2,4ꢀCl2
2,6ꢀCl2
34
0
nd
33
20
28
89
48
28
0
nd
1j
65
14
37
14
5
2,500
66
1k
1l
57
22
13
1m
16
7,800
1n
3,4ꢀCl2
3,5ꢀCl2
90
88
40
46
71
12
83
40
92
83
50
32
71
270
219
4
(AC-1045)
1o
1p
3ꢀCF3,4ꢀCl
100
100
(AC-3099)
26
38
90
8
ndb
nd
11
32
63
0
nd
nd
nd
1u
1v
ꢀ
ꢀ
ꢀ
19
36
16
19
156
1w
99
36
115
42
13
345
1x
ꢀ
3H]ꢀFFRFꢀNH2 was used as radioligand for competition assays with both hNPFF1 & 2 receptors; a Values are means of two experiments
performed in duplicates; b nd : not determined. c na: not active, Ki > 5 µM
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 33
1
2
3
Table 2 Heteroaryl AGH derivatives and their binding affinities for NPFF1R & NPFF2R
4
5
.
6
7
8
9
NPFF1-R
NPFF2-R
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
% inhibition at
% inhibition at
Ki
Ki
(nM)a
(nM)a
nº
Het
5µM
0.5µM
5µM
0.5µM
X
H
1q
83
14
93
185
32
92
0
1,510
1r
100
7
52
4
(AC-2616)
2ꢀCF3
1s
1t
74
98
7
na
83
26
54
0
0
na
2ꢀNH2
3,4ꢀCl2
26
2,350
87
21
nd
38
13
nd
1y
32
47
0
0
nd
nd
5
0
nd
nd
1z
1aa
68
31
98
32
47
22
84
nd
46
30
10
0
nd
nd
1ab
1ac
[3H]ꢀFFRFꢀNH2 was used as radioligand for competition assays with both hNPFF1 & 2 receptors; aValues are means of two
experiments performed in duplicates; b nd: not determined.
As firstly introduced by Acadia researchers, the 5ꢀarylfuranꢀ2ꢀyl aminoguanidine hydrazones
1q and 1r (table 2) appeared as another promising chemical subꢀseries. However, no binding
29
data were available for SAR analysis. The specific role played by the furan ring remains
unclear. Again, aromatic substitution led to compounds with significantly increased affinities.
In particular the 2ꢀCF3 derivative 1r (AC-2616) showed a strong affinity towards both
ACS Paragon Plus Environment

Page 13 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
NPFFR subtypes. The other compounds listed in Table 1 are homologues (1n, 1v), naphtalen
(1w, 1x), and heterocyclic derivatives (1yꢀ1ac), to be compared with 1a. They were found
poorly active, except the 3ꢀindolyl compound 1ab, which presented a binding profile similar
to that of 1a (Ki = 84 nM, NPFF1R).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3: Nꢀsubstituted aminoguanidine hydrazones and their affinities for NPFF1R and
R1
NPFF2R
H
N
NHR
NH
N
NH2
NH
N
N
X
X
8
4
NPFF1-R
% inhibition
at
NPFF2-R
% inhibition
at
5µM 0.5µM
Ki
(nM)
Ki
(nM)
Yield
(%)
69
79
91
70
nº
X
R (or R1)
Method
5µM 0.5µM
3ꢀCl
3ꢀCl
3ꢀCl
3ꢀCl
Me
Ph
Bn
A
B
A
A
27
42
52
50
28
28
18
11
nda
nd
nd
nd
64
26
34
49
25
12
12
8
nd
nd
nd
nd
4a
4b
4c
4d
(CH2)2Ph
3,4ꢀCl2
3,4ꢀCl2
3,4ꢀCl2
3,4ꢀCl2
3,4ꢀCl2
Me
Ph
A
B
92
57
78
48
70
76
78
82
57
41
51
27
13
26
14
13
21
27
8
nd
nd
nd
nd
nd
nd
nd
nd
nd
81
27
45
56
30
59
32
16
23
47
0
nd
nd
nd
nd
nd
nd
nd
nd
nd
4e
4f
Bn
A/B
A
71/59
72
13
3
4g
4h
4i
(CH2)2Ph
(CH2)3Ph
A
58
0
3,4ꢀCl2 Nꢀbenzylpiperidinyl
3,4ꢀCl2 Nꢀbenzylpiperazino
A
A
A
B
75
11
5
4j
60
4k
8a
8b
3,4ꢀCl2
3,4ꢀCl2
Me
Bn
80
5
92
14
3
Further substitutions either at the terminal nitrogen (cpds 4, R ≠ H), or nonꢀterminal guanidine
nitrogen (cpds 8, R1 ≠ H), led to inactive compounds (table 3). It may highlight some critical
interactions involving the whole aminoguanidine fragment present in AGH compounds 1, 4
and 8. This hypothesis was further supported by inactivity of imidazoline 12 and
tetrahydropyrimidine 13 (table 4). The last mode of rigidification, (type VI, Fig. 2) is new and
particularly efficient, as illustrated by data in Table 5.
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 33
1
2
3
Table 4. Affinities of cyclic AGH’S 12-13
4
5
6
7
8
NPFF1-R
% inhibition at
NPFF2-R
% inhibition at
Ki
(nM)
Ki
(nM)
nº
12
13
5µM
34
0.5µM
12
5µM
18
0.5µM
9
nda
nd
0
nd
nd
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
41
19
39
11
[3H]ꢀFFRFꢀNH2 was used as radioligand for competition assays
with both hNPFF1 & 2 receptors; a nd: not determined.
The resulting dihydrotriazines 16 showed potent affinities towards NPFF receptors. The
unsubstituted derivative 16a showed a Ki of less than 20 nM on NPFF1R with a good
selectivity profile, whereas introduction of two chlorine atoms in 3ꢀ and 4ꢀpositions (cpd
16b), restored a good affinity for NPFF2R besides a similar affinity for NPFF1R (Ki = 29 nM
and 44 nM on NPFF1R and NPFF2R, respectively). Other series in Table 5 could be
considered as two different types of bioisosteres of compounds 16. The first series (cpds 14,
X= N, Y= S) was found inactive, whereas the second series (cpds 22, X= C, Y= NH) was
again more promising with highly potent representatives
.
The unsubstituted carbaisostere (dihydropyrimidine 22a) showed a moderate affinity for both
receptor subtypes with Ki in the 100 nM range. Among the different substitutions introduced
onto the phenyl ring of 22a, beneficial chlorine effects were again particularly interesting.
Chlorination in 4ꢀposition (cpd 22c) led to a potent (Ki = 1.3nM) and selective (2 orders of
magnitude) ligand of NPFF1ꢀR. Introduction of a second chlorine atom in 3ꢀposition did not
modify the affinity towards NPFF1ꢀR, but restored a good affinity for NPFF2R (22e, Ki =
0.91 nM and 37 nM for NPFF1R and NPFF2R, respectively). Finally, as observed in the
AGH series, substitution at the exocyclic nitrogen led to compounds with strongly decrease
(cpd 25a), or loss of affinity (cpds 25b, c).
ACS Paragon Plus Environment

Page 15 of 33
ACS Chemical Neuroscience
1
2
3
Table 5. 6ꢀaryl dihydroꢀ3ꢀaminotriazines and their isosteres
4
5
6
7
8
9
NPFF1-R
NPFF2-R
% inhibition at
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
% inhibition at
Ki (nM)a
Ki (nM)a
nº
X
N
Y
S
R1
H
R2
5µM
12
0.5µM
0
5µM
0
0.5µM
0
14a
14b
14c
14d
14e
16a
16b
22a
22b
22c
22d
22e
25a
25b
25c
H
nab
na
na
na
N
S
3ꢀCl
3,4ꢀCl2
H
H
0
21
21
20
72
64
78
74
84
93
100
65
27
95
0
0
0
N
S
H
0
na
15
2
0
na
N
S
Ph
16
0
na
0
na
N
S
H
Bn
na
14
18
70
47
21
41
48
93
32
36
56
14
0
na
N
NH
NH
NH
NH
NH
NH
NH
NH
NH
NH
H
H
30
54
58
33
28
45
85
28
30
21
16
29
150
74
1.3
21
1
na
N
3,4ꢀCl2
H
H
52
24
6
48
CH
CH
CH
CH
CH
CH
CH
CH
H
69
3ꢀCl
4ꢀCl
4ꢀOMe
3,4ꢀCl2
3,4ꢀCl2
3,4ꢀCl2
3,4ꢀCl2
H
2,740
260
210
37
H
5
H
H
6
65
20
33
16
Bn
60
na
430
na
nꢀPr
(CH2)4ꢀPh
1563
358
[3H]ꢀFFRFꢀNH2 was used as radioligand for competition assays with both hNPFF1 & 2 receptors; a Values are means of at
least two experiments performed in duplicates; bna: not active, Ki > 5 µM
Some other structural analogues resembling the structure of active azines 16, 22 were
synthesized (for preparation see SI) and evaluated for their affinity toward NPFF1R and
NPFF2R (Figure 3). All of them showed low or no affinity towards both NPFF receptor
subtypes. This inactivity may result from electronic considerations emphasizing i) the lack of
NH group within the ring system (aromatic azines 20b and 26). ii) the presence of an NH
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 33
1
2
3
4
5
group within a strongly basic cyclic guanidine (cpd 27). iii) some critical steric hindrance (cpd
28).
6
7
8
Figure 3. Structurally analogs of compound 20, 22 and their binding affinity for NPFF1R and
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NPFF2R
The compounds earlier synthesized by Acadia researchers (ACꢀseries) and our most
promising hits from new chemical series (16a-b, 22c and 22e) were selected for further
pharmacological evaluation. In a first step, we confirmed affinities of 1n (AC-1045), 1p (AC-
3099), 1r (AC-2616), 16a, 16b, 22c and 22e for both NPFF1R and NPFF2R by using [125I]ꢀ
1DMeꢀNPFF as radioligand (see SI, table1). We further evaluated the selectivity of these
compounds toward the other RFꢀamide neuropeptide receptors PrRPR (alias GPR 10), Kiss1R
(alias GPR54) and QRFPR (alias GPR103) as illustrated in Table 6. Except 1p (AC-3099)
that displayed an affinity between 0,5µM and 5µM on QRFPR, all other compounds had Ki
values that were over 5 µM (or even over 20 µM for some compounds on PrRPR and Kiss1R)
for the three RFꢀamide receptors. Altogether these data indicate that the selected compounds
display a good selectivity for NPFF receptors.
ACS Paragon Plus Environment

Page 17 of 33
ACS Chemical Neuroscience
1
2
3
Table 6. Affinity of the most relevant NPFFR ligands for PrRP, Kiss1 and QRFP receptors.
4
5
6
7
8
9
Ki PrRPR Ki Kissꢀ1R Ki QRFPR
N°
Code29
(µM)a
± 5
(µM)a
> 20
(µM)a
> 5
1n
1r
AC-1045
AC-2616
AC-3099
> 20
> 5
> 20
> 5
± 5
0.5 > Ki > 5
> 5
1p
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
16a
16b
22c
22e
> 5
> 5
> 5
> 5
> 5
> 20
> 20
> 5
± 5
> 20
± 5
[3H]ꢀPrRPꢀ20, [125I]ꢀKpꢀ10 and [125I]ꢀ43RFa were used as radioligand
for competition assays with PrRP, Kiss1 and QRFP receptors (respectively);
aValues are results of at least two experiments performed in duplicates.
Examination of in vitro functional activity of the selected compounds
We then examined the agonist or antagonist activity of the seven selected compounds at both
NPFF1R and NPFF2R with HEK293ꢀhNPFF1R/NPFF2R GloSensorTm stable cell lines. We
first observed that all the selected compounds displayed a nonꢀspecific effect on cAMP
accumulation in HEKꢀGlo cells lacking recombinant hNPFF1R or hNPFF2R, when used at
concentration exceeding 10 µM (data not shown). We therefore tested these seven compounds
at 1 and 10 µM on both NPFF receptor subtypes. As shown in Figure 4A, RFRPꢀ3, the
endogenous agonist of NPFF1R, inhibited forskolinꢀstimulated production of cAMP in
hNPFF1R expressing cells while none of the seven evaluated compounds displayed agonist
activity in this assay. When tested in combination with RFRPꢀ3 (antagonist mode; Fig. 1B),
10 µM of all the compounds completely blocked the RFRPꢀ3 inhibition of forskolinꢀ
stimulated production of cAMP. Moreover, 1 µM of compounds 16a, 16b, 22c and 22e
displayed a similar effect indicating that they are more potent than compounds 1n (AC-1045),
1p (AC-3099) and 1r (AC-2616). These data clearly demonstrate that the seven selected
compounds display antagonist activity at hNPFF1R in vitro. When tested on cells expressing
NPFF2R in both agonist (Fig. 4C) and antagonist mode (Fig. 4D), except compounds 1r that
showed a significant agonist activity at 10 µM, all the others compounds displayed neither
agonist nor antagonist activity on this receptor up to 10 µM. These results contrast with
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 33
1
2
3
4
5
6
7
8
9
previous data on 1n and 1p (compound 1 and 9) 29, reporting that these two compounds were
full NPFF2R agonists with no activity (agonist or antagonist) or partial agonist activity at
NPFF1R, respectively. Although we cannot exclude that in our assay 1n and 1p could display
agonist activity at NPFF2R at concentrations higher than 10 µM, we clearly show that these
two compounds display antagonist activity at NPFF1R. This discrepancy could rely on the Rꢀ
SAT assay that was originally used to characterize these compounds as compared to more
classical radioligand binding and cAMP assays we used in this study. In conclusion, the
different compounds we characterized here, all bind to both NPFF receptor subtypes, all
display antagonist activity at NPFF1R while we detected significant agonist activity at
NPFF2R only for compound 1r (AC 2616).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 19 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Figure 4. Evaluation of the activity of seven selected compounds on NPFF1 and NPFF2
receptors. Forskolinꢀinduced cAMP accumulation was recorded in the presence of the seven
selected compounds on NPFF1R and NPFF2R stably expressed in HEKꢀGlo recombinant
cells, in the agonist (A and C, respectively) and antagonist (B and D, respectively) modes.
RFRPꢀ3 (50 nM) inhibited ≈40% of the forskolin ꢀstimulated cAMP production in HEKꢀ
NPFF1RꢀGlo cells while NPFF (50 nM) inhibited ≈30% of the forskolinꢀstimulated cAMP
production in HEKꢀNPFF2RꢀGlo cells. Antagonist activity of the seven compounds was
evaluated in presence of the 50 nM RFRPꢀ3 or NPFF as indicated in B and D. Error bars
represent the mean ± SEM of data from at least two experiments performed in duplicate.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 33
1
2
3
Cpd 1n and cpd 22e prevent long lasting hyperalgesia induced by fentanyl in rodents
4
5
In a previous work, we have shown that the pharmacological blockade of the two NPFF
6
7
8
9
receptors (NPFF1R and NPFF2R) by RF9 prevents the development of opioidꢀinduced
hyperalgesia in rodents.22,
In this study, we therefore wanted to investigate the
24
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
consequences of NPFF1R blockade with 1n (AC-1045) and 22e in two different rodent
species. The activity of these compounds was evaluated in a model of hyperalgesia induced
by acute administration of fentanyl in rats and mice, respectively.24, 17 RF9 was also tested in
these experiments as a positive control. As shown in Figure 5, fentanyl (4x80 µg/kg, sc.)
promoted a short lasting analgesic response in rats (modulation of paw pressure), which
reached a maximum at 1 h after the last injection of fentanyl and returned to basal level within
5 h. On the next days following fentanyl injection, rats exhibited a delayed hyperalgesic
response that lasted for four days.
As expected, RF9 (0.5 mg/kg) almost blocked fentanylꢀinduced hyperalgesia (HI: 100 ± 10
for fentanyl vs 0.5 ± 0.1 for RF9, F(2, 25) = 55 ; p < 0.001, one way ANOVA followed by
Fisher’s PLSD test p < 0.01). Similarly, pretreatment with compound 1n (0.5 mg/kg, sc.)
before fentanyl had no effect on short lasting fentanyl analgesia but significantly attenuated
the long lasting fentanylꢀinduced hyperalgesia (HI: 100 ± 10 for fentanyl vs 9.5 ± 5 for
compound 1n, F(2, 25) = 55 ; p < 0.001, one way ANOVA followed by Fisher’s PLSD test p
< 0.001) .
ACS Paragon Plus Environment

Page 21 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Figure 5: Effect of compound 1n on analgesic and hyperalgesic responses of rats to
fentanyl. On day 0, a single dose (0.5 mg/kg, sc.) of cpd 1n or RF9 was administrated 20 min
before four consecutive fentanyl injections (80 µg/kg, sc ; 15 min interval). Nociceptive
threshold of the animals was measured using the paw pressure test once daily before
experiment start (dꢀ2, dꢀ1, d0) to verify baseline stability, at 1h, 3h and 5h after the last
fentanyl injection to monitor analgesia amplitude (d0), and once daily (d1 to d6) to follow the
development of hyperalgesia. Hyperalgesia indexes (HI) were calculated as described in the
Data and statistical analyses section and presented as histograms. Data are expressed as mean
± S.E.M, n = 8ꢀ10. ***: p < 0.001 by Fisher’s test as compared with the fentanylꢀpretreated
saline group. ns : p > 0.05, by Fisher’s test as compared with the fentanylꢀpretreated R9 group
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We next examined the effect of compound 22e on fentanylꢀinduced analgesia and
hyperalgesia in mice (tail withdrawal latency, Fig. 6). Very similarly to what we observed in
rats, injection of fentanyl (4x60 µg/kg, sc.) displayed a strong analgesic effect followed by an
hyperalgesic state that was significantly attenuated by RF9 (5 mg/kg, sc; HI: 100 ± 10 for
fentanyl vs 32 ± 9 for RF9, F(2, 27) = 10 ; p < 0.001, one way ANOVA followed by Fisher’s
PLSD test p < 0.001). Pretreatement with compound 22e at the same dose (5 mg/kg, sc.) than
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 33
1
2
3
4
5
6
7
8
RF9 30 min before fentanyl also significantly attenuated the development of hyperalgesia
(HI: 100 ± 10 for fentanyl vs 52 ± 12 for cpd 22e, F(2, 27) = 10 ; p < 0.001, one way ANOVA
followed by Fisher’s PLSD test p < 0.01).
9
Figure 6: Effect of compound 22e on analgesic and hyperalgesic responses of mice to
fentanyl. On day 0, a single dose (5 mg/kg; sc.) of cpd 22e or RF9 was administrated to mice
20 min before fentanyl injections (4x60µg/kg, sc; 15 min interval). Nociceptive response of
the animals was measured using the tail immersion test (48°C) once daily before experiment
start (dꢀ2, dꢀ1, d0) to verify baseline stability, every 1 h after the last fentanyl injection until
return to baseline and were repeated once daily from d1 to d4. Hyperalgesia indexes (HI, %)
were calculated as described in the Data and statistical analyses section and presented as
histograms. Data are expressed as mean ± S.E.M, n = 10. **: p < 0.01, ***: p < 0.001 by
Fisher’s test as compared with the fentanylꢀpretreated saline group. ns :p > 0.05 by Fisher’s
test as compared with the fentanylꢀpretreated RF9 group.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Altogether, these results indicate that targeting NPFF receptors by AGH 1n or its
rigidified analog 22e efficiently prevents the development of hyperalgesia induced by acute
fentanyl administration in rodents. Conversely to what was originally proposed for compound
1n described as a selective NPFF2R agonist, 29 they further suggest that the antiꢀhyperalgesic
ACS Paragon Plus Environment

Page 23 of 33
ACS Chemical Neuroscience
1
2
3
activity of these compounds may not only rely on their NPFF2R character but more likely on
4
5
their capacity to efficiently block NPFF1R. This hypothesis is also supported by recent data
6
7
8
26
obtained with a compound showing clearly NPFF1R selectivity over NPFF2R.
In the
9
future, a better understanding of the respective role of both NPFF receptor subtypes in the
modulation of nociception and opioid side effects will involve the development of highly
selective compounds for these two NPFF receptor subtypes with a clear in vitro and in vivo
antagonistic profile. In conclusion combination of opioid analgesics (morphine, fentanyl,
remifentanil)43 with NPFF receptor antagonists may help to reduce adverse effects associated
with chronic opioid treatments, including OIH and analgesic tolerance, in many clinical
situations including postoperative, traumatic or tumor pain therapy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Materials and Methods
Compound synthesis and binding data using [125I]ꢀ1DMeꢀ NPFF versus [3H]ꢀFFRFꢀNH2 for
cpds 1a, 1p, 1r, 16a-b, 22c, 22e.
See supporting information
Biological evaluation of compounds
Material
Fentanyl citrate, forskolin, 3ꢀisobutylꢀ1ꢀmethylxanthine (IBMX) and poly(ethyleneimine)
solution (PEI) were from SigmaꢀAldrich (Saint Quentin Fallavier, France). [3H]ꢀFFRFꢀamide
(13.6 Ci/mmol) was from the CEA (Saclay, France). [3H]ꢀPrRPꢀ20 (150 Ci/mmol) was from
Hartmann Analytic (Braunschweig, Germany). [125I]ꢀ(DꢀTyr1, NꢀMeꢀPhe3)ꢀNPFF ([125I]ꢀ
1DMeꢀNPFF; 2200 Ci/mmol), [125I]ꢀKpꢀ10 (2200 Ci/mmol), [125I]ꢀQRFP43 (2200 Ci/mmol)
were purchased from Perkin Elmer Life and Analytical Sciences (Courtaboeuf, France).
Human RFꢀamide peptides were from Polypeptide (Strasbourg, France; Kpꢀ10, NPFF, PrRPꢀ
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 33
1
2
3
4
5
6
7
8
9
20), Peptanova (Sandhausen, Germany, 43RFa) and TebuꢀBio (Le PerrayꢀenꢀYvelines, France
; 26RFa and RFRPꢀ3). Dꢀluciferine was purchased from Synchem (FelsbergꢀAltenberg,
Germany). NꢀadamantanecarbonylꢀArgꢀPheꢀNH2 trifluoroacetate (RF9) was synthesized as
reported.22
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Receptor cDNA constructs, cell expression and membrane preparations
Human NPFF1R, NPFF2R and Kiss1R cDNAs were subcloned into the pCDNA3.1
expression vector (Invitrogen, Cergy Pontoise, France) and transfected into CHO cells before
selection for stable expression, as reported.42 CHO cells expressing human PrRPR and
QRFPR were a gift from M. Parmentier (IRIBHM, Brussells, Belgium), and were grown in
DMEM/Ham’s Fꢀ12 Nutrient Mixture (Gibco, Fisher Scientific, Villebon sur Yvette)
supplemented with 5% fetal calf serum, penicilline (100 µg/L) and streptomycine (100 U/L).
HEK293 cells stably expressing the cAMP GloSensorTM were a gift from J. Hanson
(Molecular Pharmacology GIGAꢀSignal Transduction and Medicinal Chemistry Drug
Research Centre, Liege, Belgium) and were grown in DMEM+4,5g/L Glucose (Gibco, Fisher
Scientific, Villebon sur Yvette) supplemented with 10% fetal calf serum, penicilline (100
µg/L), streptomycine (100 U/L) and hygromycine (200 µg/mL). pCDNA3 constructs
encoding human NPFF1R, NPFF2R were further transfected into these cells before G418
selection (500 µg/mL) for stable receptor expression. Membranes from CHO cells expressing
RFꢀamide receptors were prepared as reported42 and stored at ꢀ80°C before use.
Radioligand binding assays
Binding assay conditions were essentially as described previously.42 Briefly, NPFF1Rꢀ or
NPFF2Rꢀcontaining membranes were incubated with 10 nM or 3 nM [3H]ꢀFFRFꢀNH2,
respectively. When [125I]ꢀ1DMeꢀNFF was used as a radiotracer, both NPFF1Rꢀ or NPFF2Rꢀ
containing membranes were incubated with 0.015 nM of this radioligand. Nonꢀspecific
binding levels were determined in the presence of 1 µM RFRPꢀ3 (NPFF1R) or 1 µM NPFF
ACS Paragon Plus Environment

Page 25 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
(NPFF2R). [3H]ꢀPrRPꢀ20 (0.3 nM), [125I]ꢀKpꢀ10 (0.02 nM) and [125I]ꢀ43RFa (0.03 nM) were
used to label PrRP, Kiss1 and QRFP receptors (respectively) and non specific binding levels
were evaluated in the presence of 1 µM PrRPꢀ20, Kpꢀ10 and 26RFa (respectively).
Competitionꢀtype experiments were performed at 25°C, under binding equilibrium conditions
(30 min, 0.5 mL final volume for NPFF1R and NPFF2R; 1 h, 0.25 mL final volume for the
three other receptors), in the presence of various concentrations of the compounds to be tested
for their binding affinity. Membraneꢀbound radioactivity was separated from free radioligand
by rapid filtration through 96ꢀwell GF/B unifilters presoaked in 0.1% PEI for [3H]ꢀPrRPꢀ20,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[
125I]ꢀKpꢀ10 and [125I]ꢀ43RFa and 0.5% PEI for [125I]ꢀ1DMeꢀNPFF using a cell harvester
(Perkin Elmer Life and Analytical Sciences, Courtaboeuf, France) and quantified using a
TopCount scintillation counter (Perkin Elmer).
Glo Sensor cAMP assay
The functionnal characterization of the selected compounds was performed by Glo Sensor
cAMP kinetic assay, a validated approach for measuring Gαi/s activation by GPCRs
(DiRaddo et al., 2014; Gilissen et al., 2015).44ꢀ45 HEKꢀGlo cells expressing either NPFF1 or
NPFF2 receptors were suspended (106 cells per mL) in physiological Hepes buffer (10mM
HEPES, 0.4mM NaH2PO4, 137.5mM NaCl, 1.25mM MgCl2, 1.25mM CaCl2, 6mM KCl,
10mM glucose and 1mg/mL bovine serum albumin, pH 7.4) supplemented with 1mM Dꢀ
Luciferine. After equilibration for 2h at 25°C, luminescence levels were recorded in real time
in 96ꢀwell plates using a FlexstationR 3 (Molecular Devices, Sunnyvale CA, USA). Nonꢀ
specific effect of the different compounds was evaluated in HEKꢀGlo cells without any
recombinant human RFꢀamide receptors. To test the agonist activity on NPFF1R and
NPFF2R, compounds were injected 15 minutes before addition of forskolin (0,5µM) and
readings were pursued for 90 minutes. To evaluate their antagonist properties, compounds
were preꢀincubated with cells for 15 minutes before prototypical agonists for NPFF1R
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 33
1
2
3
4
5
6
7
8
9
(RFRPꢀ3) and NPFF2R (NPFF) receptors. According to their preferential coupling to Gi/o
proteins, the stimulation of NPFF1R and NPFF2R by agonists was monitored as a doseꢀ
dependent reduction in steadyꢀstate luminescence levels, reflecting the inhibition of forskolinꢀ
induced cAMP accumulation. Experiments were performed at 25°C in the presence of 0,5mM
IBMX to prevent the degradation of cAMP by phosphodiesterases.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Measurement of the nociceptive threshold in rats and mice
The nociceptive mechanical response of rats was measured using the pawꢀpressure
vocalization test adapted from the Randall–Selitto method (Elhabazi et al., 2014; Kayser et
al., 1990).46ꢀ47 Uniformly increasing pressure (in g) was applied to the hind paw until the rat
squeaked. The Basile analgesimeter (Apelex, Massy, France; 1 mm stylus tip diameter) was
used and a 600 g cutꢀoff value was set to prevent tissue damage.
The nociceptive thermal response of mice was determined using the tail immersion test as
previously described (Elhabazi et al., 2014; Simonin et al., 1998).47ꢀ48 Briefly, mice were
restrained in a grid pocket and their tail was immersed in a thermostated water bath. The
latency (in sec) for tail withdrawal from hot water (48 ± 0.5 °C) was taken as a measure of the
nociceptive response. In the absence of any nociceptive reaction, a cutꢀoff value of 25 sec was
set to avoid tissue damage.
Nociception tests were performed on male SpragueꢀDawley rats (250ꢀ350 g weight; Charles
River Laboratories, L’Arbresle, France) and on C57BL/6N male mice (25ꢀ30 g weight;
Taconic, Denmark). Animals were housed in groups of three to five per cage and kept under a
12 h/12 h light/dark cycle at 21 ± 1°C with ad libitum access to food and water and were
habituated to the testing room and equipment and handled for 2 weeks before starting
behavioural experiments.
All experiments were carried out in accordance with the European directive for the care of
laboratory animals (2010/63/EU) and approved by the local committee (CREMEAS).
ACS Paragon Plus Environment

Page 27 of 33
ACS Chemical Neuroscience
1
2
3
4
5
Data and statistical analyses
6
7
8
9
Binding and functional data were analyzed using Prism 4.0 (GraphPad Software, San Diego,
CA, USA).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Data for nociceptive tests are expressed as mean values ± S.E.M. for 8ꢀ10 animals per group.
Hyperalgesia was quantified as the Hyperalgesia Index (HI) in the fentanyl inducedꢀ
hyperalgesia experiment, which represents the area above the curve expressed as a mean
percentage relative to the reference values for the fentanylꢀpreatreated saline group. Data were
analyzed using oneꢀway analysis of variance (ANOVA). Postꢀhoc analyses were performed
with Fisher’s PLSD test. The level of significance was set at p < 0.05. All statistical analyses
were carried out using the StatView software.
Associated Content
Supporting Informations:
ꢀ Compound synthesis and analytical characterization of compounds 1-28
ꢀ Binding data using [125I]ꢀ1DMeꢀ NPFF versus [3H]ꢀFFRFꢀNH2 for cpds 1a, 1p, 1r, 16a-b,
22c, 22e.
Author Contributions
Contact information: Concerning medicinal chemistry: Martine Schmitt, phone +33
0368854231, mschmitt@unistra.fr ; concerning pharmacological evaluation: Frédéric
Simonin, phone ++33 368854875, simonin@unistra.fr
# These authors participated equally to this work
ACKNOWLEDGEMENTS
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 33
1
2
3
4
5
6
7
8
9
We thank Dr M. Parmentier (IRIBHM, Brussels, Belgium) for the gift of CHO cells
expressing human PrRPR and QRFPR. This work also benefited from the efficient support of
the Integrative Biological Chemistry Platform (UMS 3286 CNRS, Illkirch, France). We thank
the staff from the neurobiology and behavior service at Institut Clinique de la Souris (Illkirch,
France) for technical assistance and advices. This work was supported by the CNRS,
Université de Strasbourg, Alsace BioValley and by grants from SATT Conectus, Agence
National de la Recherche (ANR 08 EBIO 014.02, ANRꢀ13ꢀBSV1ꢀ0001), Conseil Régional
d’Alsace (Pharmadol), Communauté Urbaine de Strasbourg (Pharmadol), ICFRC
(Pharmadol), OSEO (Pharmadol), Direction Générale des Entreprises (Pharmadol). This work
has been published within the LABEX ANRꢀ10ꢀLABXꢀ0034_Medalis and received financial
support from the French government managed by "Agence National de la Recherche" under
"Programme d'investissement d'avenir”.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Notes
The authors declare no competing financial interest
RERERENCES
(1) Yang, H. Y., Fratta, W., Majane, E. A., and Costa, E. (1985) Isolation, sequencing,
synthesis, and pharmacological characterization of two brain neuropeptides that modulate the
action of morphine. Proc. Natl. Acad. Sci. 82, 7757–7761.
(2) Perry, S. J., YiꢀKung Huang, E., Cronk, D., Bagust, J., Sharma, R., Walker, R. J., Wilson,
S., and Burke, J. F. (1997) A human gene encoding morphine modulating peptides related to
NPFF and FMRFamide. FEBS Lett. 409, 426–430.
(3) Bonini, J. A., Jones, K. A., Adham, N., Forray, C., Artymyshyn, R., Durkin, M. M.,
Smith, K. E., Tamm, J. A., Boteju, L. W., Lakhlani, P. P., Raddatz, R., Yao, W.ꢀJ., Ogozalek,
K. L., Boyle, N., Kouranova, E. V., Quan, Y., Vaysse, P. J., Wetzel, J. M., Branchek, T. A.,
Gerald, C., and Borowsky, B. (2000) Identification and Characterization of Two G Proteinꢀ
coupled Receptors for Neuropeptide FF. J. Biol. Chem. 275, 39324–39331.
(4) Quillet, R., Ayachi, S., Bihel, F., Elhabazi, K., Ilien, B., and Simonin, F. (2016) RFꢀamide
neuropeptides and their receptors in Mammals: Pharmacological properties, drug
development and main physiological functions. Pharmacol. Ther. 160, 84–132.
ACS Paragon Plus Environment

Page 29 of 33
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
(5) Gouardères, C., Quelven, I., Mollereau, C., Mazarguil, H., Rice, S. Q. ., and Zajac, J. .
(2002) Quantitative autoradiographic distribution of NPFF1 neuropeptide FF receptor in the
rat brain and comparison with NPFF2 receptor by using [125I]YVP and [125I]EYF as
selective radioligands. Neuroscience 115, 349–361.
(6) Murase, T., Arima, H., Kondo, K., and Oiso, Y. (1996) Neuropeptide FF reduces food
intake in rats. Peptides 17, 353–354.
(7) Cline, M. A., and Mathews, D. S. (2008) Anoretic effects of neuropeptide FF are mediated
via central mu and kappa subtypes of opioid receptors and receptor ligands. Gen. Comp.
Endocrinol. 159, 125–129.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(8) Maletínská, L., Tichá, A., Nagelová, V., Špolcová, A., Blechová, M., Elbert, T., and
Železná, B. (2013) Neuropeptide FF analog RF9 is not an antagonist of NPFF receptor and
decreases food intake in mice after its central and peripheral administration. Brain Res. 1498,
33–40.
(9) Kavaliers, M., and Colwell, D. D. (1993) Neuropeptide FF (FLQPQRFamide) and IgG
from neuropeptide FF antiserum affect spatial learning in mice. Neurosci. Lett. 157, 75–78.
(10) Lénárd, L., Kovács, A., Ollmann, T., Péczely, L., Zagoracz, O., Gálosi, R., and László,
K. (2014) Positive reinforcing effects of RFamideꢀrelated peptideꢀ1 in the rat central nucleus
of amygdala. Behav. Brain Res. 275, 101–106.
(11) Kovács, A., László, K., Zagoracz, O., Ollmann, T., Péczely, L., Gálosi, R., and Lénárd,
L. (2017) Effects of RFamideꢀrelated peptideꢀ1 (RFRPꢀ1) microinjections into the central
nucleus of amygdala on passive avoidance learning in rats. Neuropeptides 62, 81–86.
(12) Allard, M., Labrouche, S., Nosjean, A., and Laguzzi, R. (1995) Mechanisms underlying
the cardiovascular responses to peripheral administration of NPFF in the rat. J. Pharmacol.
Exp. Ther.407, 245ꢀ255.
(13) Arima, H., Murase, T., Kondo, K., Iwasaki, Y., and Oslo, Y. (1996) Centrally
administered neuropeptide FF inhibits arginine vasopressin release in conscious rats.
Endocrinology. 137, 1523ꢀ1529.
(14) Ayachi, S., and Simonin. (2014) Involvement of Mammalian RFꢀAmide Peptides and
Their Receptors in the Modulation of Nociception in Rodents. Front. Endocrinol. 5, 158.
(15) Rothman, R. B. (1992) A review of the role of antiꢀopioid peptides in morphine tolerance
and dependence. Synapse 12, 129–138.
(16) Melzack, R. The stragedy of needless pain. Sci Am. 262 (2). 27‑33.
(17) Collin, E., Poulain, P., Gauvain ꢀPiquard, A., Petit, G., and PichardꢀLeandri, E. (1993) Is
disease progression the major factor in morphine “tolerance” in cancer pain treatment. Pain.
(18) McNally, G. P. (1999) Pain facilitatory circuits in the mammalian central nervous
system: their behavioral significance and role in morphine analgesic tolerance. Neurosci.
Biobehav. Rev. 23, 1059–1078.
(19) Célèrier, E., Rivat, C., Jun, Y., Laulin, J.ꢀP., Larcher, A., Reynier, P., and Simonnet, G.
(2000) Longꢀlasting Hyperalgesia Induced by Fentanyl in Rats: Preventive Effect of
Ketamine. Anesthesiology 92, 465ꢀ472.
(20) Colvin, L. A., and Fallon, M. T. (2010) Opioidꢀinduced hyperalgesia: a clinical
challenge. Br. J. Anaesth. 104, 125–127.
(21) Angst, M. S., and Clark, J. D. (2006) Opioidꢀinduced Hyperalgesia: A Qualitative
Systematic Review. Anesthesiology 104, 570–587.
(22) Simonin, F., Schmitt, M., Laulin, J.ꢀP., Laboureyras, E., Jhamandas, J. H., MacTavish,
D., Matifas, A., Mollereau, C., Laurent, P., Parmentier, M., Kieffer, B. L., Bourguignon, J.ꢀJ.,
and Simonnet, G. (2006) RF9, a potent and selective neuropeptide FF receptor antagonist,
prevents opioidꢀinduced tolerance associated with hyperalgesia. Proc. Natl. Acad. Sci. 103,
466–471.
ACS Paragon Plus Environment